Viruses have left no fossil record, hence we must rely on scrutiny of existing viruses for clues to viral origins and evolution. Much can be learned by comparing the nucleotide sequences of the thousands of viral and cellular genes now available in the burgeoning international computer data banks. F Fowever, it must be recognized that we are looking through a tiny window ol time in the context of the history of life on earth. Although sequencing the genomes of contemporary viruses can shed light on relationships between (hem and certain cellular genes, we are still in the dark about the origin ol viruses Some scientists regard it as axiomatic that viruses evolved originally from DNA or RNA already present in a cellular organelle or chromosome, or from some form of intracellular parasite such as a bacterium, whereas others picture a primeval "RNA world" in which a form of self-replicating RNA akin to modern viroids predated DNA, proteins, and cells. We also have no idea whether all viruses evolved from a single progenitor, although there is evidence that the plus sense RNA viruses may have.
A computer search reveals that the genomes of virtually all plus sense RNA viruses, whether of animal or plant origin, contain an RNAdependent RNA polymerase gene with certain conserved motifs, suggesting that this essential en/yme may have been the earliest viral protein. Most families of the larger RNA viruses also carry genes for an RNA helicase and a proteinase activity, both of which display sufficient lesemblance to their cellular homo-logs to indicate that they were originally acquired from cellular nucleic acid by recombination, or vice versa Indeed, there is much evidence to suggest that among viruses genetic recombination has been a more impoitanl evolutionary mechanism than point mutation, and has been responsible for major changes such as the production of the progenitors of all higher level laxa The theory of "modular evolution" of plus sense RNA viral genomes postulates that, once the cluster of genes essential for genome replication was m place m the prototype virusfes), other genes encoding accessory and structural proteins, less vital but advantageous to the virus, were added to the genome by recombination; subsequently these "modules" or "cassettes" have been acquired from or exchanged with the genome of other, related or unrelated, viruses by genetic recombination or reassortment. Minor changes continue to occur at a very high frequency as a result of point mutations or less frequently nucleotide insertion or deletion. Such mutations do not occur at the same rate in all genes of animal viruses; in influenza A virus, for example, they are more abundant in surface proteins subject to selection by neutralizing antibodies.
Because of the importance of serological methods for diagnostic purposes, clinical virologists place great emphasis on the surface proteins, which distinguish viral strains of relatively recent origin from one another. Phyio-genetically, however, the key enzymes concerned with genome replication, as well as gene order, the nature and location of noncoding regulatory sequences, and key features of the viral replication strategy, are more significant in defining a family and its relationship to other families from which it may have diverged millions of years ago. In the so-called polythchc approach to taxonomy, viruses are grouped into genera that share a unique set of characters, some but not all of which may be shared with viruses of other genera. These laxonomically useful characters are less variable than some of the other phenotypic characters that are of greater diagnostic, pathogenetic, and epidemiologic importance.
Similar principles and algorithms have been applied to construct evolutionary trees of other viruses. The families of minus sense RNA viruses and of DNA viruses are more heterogeneous than those of the plus sense RNA viruses, and may have arisen from a number of distinct protoviruses The large DNA viruses such as poxviruses and herpesviruses appear to have captured the genes for numerous cellular enzymes from their host cells at some time in the distant past. The retroviruses share a common genetic feature, the "gag-pol rephcon" (which encodes a reverse transcriptase) with the "pararetroviruses" (such as hepadnaviruses), as well as with the various classes of noninfectious "retroelements" known as retrotiansposons, retro-posons, and retrons.
The genetic mechanisms described in this chapter, opeiating under the pressure of Darwinian selection, have been clearly incriminated in several recent important examples of viral evolution ("lable 4-1). Here we shall content ourselves with a description of two viruses that illustrate particularly well the dramatic impacl of evolution, even over a relatively short lime span. One,
Examples ol Genetic Mechanisms That have Affected Vnal Evolution"
I n tra molecu la r rccomb i na tion
Recombination and mutation Biased hypormutation''
(uridine to cytosine transitions) Genetic rearrangement'
Lethal chicken influenza due lo a single point mutation Western equine encephalitis virus produced by recombination between eastern equine encephalitis virus and a Sindbis-Iike alphavirus Pandemic human influenza A subtypes H2N2 (1937) and
H3N2 <1%8) Changes in poliovaccme following vaccination Evolution of subacute sclerosmg panencephalitis virus from measles virus Evolution of rubella virus
" Based on E D Kilbourne, Curr Opm Immunol 3, 518 (1991)
Missense mutations of M gene ' Compared with the alphavirus genome, the order of helicase and NS P3 region is reversed in rubella virus myxoma virus, highlights the importance of changes in viral virulence and host resistance The other, influenza A virus, illustrates how effectively viruses can evolve to avoid the immune response of the host.
Was this article helpful?